Congenital heart disease detection by pediatric electrocardiogram based deep learning integrated with human concepts
Abstract Early detection is critical to achieving improved treatment outcomes for child patients with congenital heart diseases (CHDs). Therefore, developing effective CHD detection techniques using low-cost and non-invasive pediatric electrocardiogram are highly desirable. We propose a deep learnin...
Main Authors: | , , , , , , , , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Nature Portfolio
2024-02-01
|
Series: | Nature Communications |
Online Access: | https://doi.org/10.1038/s41467-024-44930-y |
_version_ | 1827326916693065728 |
---|---|
author | Jintai Chen Shuai Huang Ying Zhang Qing Chang Yixiao Zhang Dantong Li Jia Qiu Lianting Hu Xiaoting Peng Yunmei Du Yunfei Gao Danny Z. Chen Abdelouahab Bellou Jian Wu Huiying Liang |
author_facet | Jintai Chen Shuai Huang Ying Zhang Qing Chang Yixiao Zhang Dantong Li Jia Qiu Lianting Hu Xiaoting Peng Yunmei Du Yunfei Gao Danny Z. Chen Abdelouahab Bellou Jian Wu Huiying Liang |
author_sort | Jintai Chen |
collection | DOAJ |
description | Abstract Early detection is critical to achieving improved treatment outcomes for child patients with congenital heart diseases (CHDs). Therefore, developing effective CHD detection techniques using low-cost and non-invasive pediatric electrocardiogram are highly desirable. We propose a deep learning approach for CHD detection, CHDdECG, which automatically extracts features from pediatric electrocardiogram and wavelet transformation characteristics, and integrates them with key human-concept features. Developed on 65,869 cases, CHDdECG achieved ROC-AUC of 0.915 and specificity of 0.881 on a real-world test set covering 12,000 cases. Additionally, on two external test sets with 7137 and 8121 cases, the overall ROC-AUC were 0.917 and 0.907 while specificities were 0.937 and 0.907. Notably, CHDdECG surpassed cardiologists in CHD detection performance comparison, and feature importance scores suggested greater influence of automatically extracted electrocardiogram features on CHD detection compared with human-concept features, implying that CHDdECG may grasp some knowledge beyond human cognition. Our study directly impacts CHD detection with pediatric electrocardiogram and demonstrates the potential of pediatric electrocardiogram for broader benefits. |
first_indexed | 2024-03-07T14:52:27Z |
format | Article |
id | doaj.art-4c34fb493748466c9c5fca70e5485f5c |
institution | Directory Open Access Journal |
issn | 2041-1723 |
language | English |
last_indexed | 2024-03-07T14:52:27Z |
publishDate | 2024-02-01 |
publisher | Nature Portfolio |
record_format | Article |
series | Nature Communications |
spelling | doaj.art-4c34fb493748466c9c5fca70e5485f5c2024-03-05T19:37:37ZengNature PortfolioNature Communications2041-17232024-02-0115111310.1038/s41467-024-44930-yCongenital heart disease detection by pediatric electrocardiogram based deep learning integrated with human conceptsJintai Chen0Shuai Huang1Ying Zhang2Qing Chang3Yixiao Zhang4Dantong Li5Jia Qiu6Lianting Hu7Xiaoting Peng8Yunmei Du9Yunfei Gao10Danny Z. Chen11Abdelouahab Bellou12Jian Wu13Huiying Liang14State Key Laboratory of Transvascular Implantation Devices of the Second Affiliated Hospital, Zhejiang University School of Medicine, Zhejiang UniversityMedical Big Data Center, Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical UniversityGuangdong Cardiovascular Institute, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical SciencesLiaoning Engineering Research Center of Intelligent Diagnosis and Treatment EcosystemLiaoning Engineering Research Center of Intelligent Diagnosis and Treatment EcosystemMedical Big Data Center, Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical UniversityGuangdong Cardiovascular Institute, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical SciencesMedical Big Data Center, Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical UniversityMedical Big Data Center, Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical UniversityCollege of Information Technology and Engineering, Guangzhou College of CommerceZhuhai Precision Medical Center, Zhuhai People’s Hospital/ Zhuhai Hospital Affiliated with Jinan University, Jinan UniversityDepartment of Computer Science and Engineering, University of Notre DameInstitute of Sciences in Emergency Medicine, Guangdong Provincial People’s Hospital, Guangdong Academy of Medical SciencesState Key Laboratory of Transvascular Implantation Devices of the Second Affiliated Hospital, Zhejiang University School of Medicine, Zhejiang UniversityMedical Big Data Center, Guangdong Provincial People’s Hospital (Guangdong Academy of Medical Sciences), Southern Medical UniversityAbstract Early detection is critical to achieving improved treatment outcomes for child patients with congenital heart diseases (CHDs). Therefore, developing effective CHD detection techniques using low-cost and non-invasive pediatric electrocardiogram are highly desirable. We propose a deep learning approach for CHD detection, CHDdECG, which automatically extracts features from pediatric electrocardiogram and wavelet transformation characteristics, and integrates them with key human-concept features. Developed on 65,869 cases, CHDdECG achieved ROC-AUC of 0.915 and specificity of 0.881 on a real-world test set covering 12,000 cases. Additionally, on two external test sets with 7137 and 8121 cases, the overall ROC-AUC were 0.917 and 0.907 while specificities were 0.937 and 0.907. Notably, CHDdECG surpassed cardiologists in CHD detection performance comparison, and feature importance scores suggested greater influence of automatically extracted electrocardiogram features on CHD detection compared with human-concept features, implying that CHDdECG may grasp some knowledge beyond human cognition. Our study directly impacts CHD detection with pediatric electrocardiogram and demonstrates the potential of pediatric electrocardiogram for broader benefits.https://doi.org/10.1038/s41467-024-44930-y |
spellingShingle | Jintai Chen Shuai Huang Ying Zhang Qing Chang Yixiao Zhang Dantong Li Jia Qiu Lianting Hu Xiaoting Peng Yunmei Du Yunfei Gao Danny Z. Chen Abdelouahab Bellou Jian Wu Huiying Liang Congenital heart disease detection by pediatric electrocardiogram based deep learning integrated with human concepts Nature Communications |
title | Congenital heart disease detection by pediatric electrocardiogram based deep learning integrated with human concepts |
title_full | Congenital heart disease detection by pediatric electrocardiogram based deep learning integrated with human concepts |
title_fullStr | Congenital heart disease detection by pediatric electrocardiogram based deep learning integrated with human concepts |
title_full_unstemmed | Congenital heart disease detection by pediatric electrocardiogram based deep learning integrated with human concepts |
title_short | Congenital heart disease detection by pediatric electrocardiogram based deep learning integrated with human concepts |
title_sort | congenital heart disease detection by pediatric electrocardiogram based deep learning integrated with human concepts |
url | https://doi.org/10.1038/s41467-024-44930-y |
work_keys_str_mv | AT jintaichen congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT shuaihuang congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT yingzhang congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT qingchang congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT yixiaozhang congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT dantongli congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT jiaqiu congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT liantinghu congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT xiaotingpeng congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT yunmeidu congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT yunfeigao congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT dannyzchen congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT abdelouahabbellou congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT jianwu congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts AT huiyingliang congenitalheartdiseasedetectionbypediatricelectrocardiogrambaseddeeplearningintegratedwithhumanconcepts |